assumed to be from an exogenous source. From 19 patients, clinical samples were taken during the same hospitalization period as the study sample was taken. For three of these 19 patients (15.9%), the S. aureus was not identical to the clinical strain, indicating a potential exogenous source. It is important to prevent acquisition of S. aureus, as noncarriers have a higher mortality following bacteremia compared to carriers (16). It is estimated that 80% of nosocomial S. aureus infections are endogenous, and approximately 20% exogenous (16, 17). This is in line with our finding that 15.9% of patients had a potential exogenous source. Another explanation for the difference between the nasal and clinical strain could be that patients carried multiple strains upon admission, of which we only identified one (11, 18). However, only 6.6-10% of S. aureus carriers carry multiple strains simultaneously. While morphologically different S. aureus from one nasal sample were analyzed, we did not identify multiple S. aureus strains in any of the nasal samples. Besides acquisition, we showed that 35.2% of patients positive for MSSA upon admission were negative at discharge. Although we did not determine the use of antibiotics, antibiotic use could explain the relatively large proportion no longer colonized or not cultivable upon discharge. Overall, our results regarding prevalence of MSSA, MRSA, and ESBL-E are in line with previous Dutch studies, but we identified that transmission within the hospital often remains unidentified. We showed that transmission in the hospital is invisible, which highlights gaps in the current surveillance methods. In total, during the study period, we observed acquisition of ESBL-E in 20 patients, and of MSSA in 15 patients, which theoretically could have been prevented. Consequently, we conclude that there is a need for improving or adjusting current surveillance methods within the hospital. Screening of patients upon admission In Chapter 2.1 we observed that 91.6% of ESBL-E carriers were unknown carriers upon admission, and remained unidentified throughout their hospitalization (4). Consequently, these patients were not cared for in isolation, increasing the chances on transmission to other patients. The majority of these patients remained unidentified carriers throughout their hospitalization. The high percentage of unidentified carriers highlights the need for effective screening methods upon admission. Currently, there is a nationally implemented universal risk assessment combined with risk-based screening upon admission to the hospital. This assessment consists of a number of questions determining the risk on being colonized with a MDRO (19). When a patient is deemed at risk, risk-based screening is performed and, depending on if the patient is considered to be at high or low risk, the patient is preemptively placed in isolation. However, van Hout et al. showed that the currently nationally installed universal risk assessment combined with risk-based screening is not effective in identifying MDRO carriers upon admission (20). They identified that the majority of carriers identified through the risk assessment were identified through the question: “are you a known carrier of a MDRO?”. Consequently, they propose abandoning the currently 4 187 Summarizing discussion
RkJQdWJsaXNoZXIy MTk4NDMw